stress testing: evaluation of different approaches for ... · stress testing introduction stress...

© 2019 Fair Isaac Corporation. Confidential. 1

© 2019 Fair Isaac Corporation. Confidential.

This presentation is provided for the recipient only and cannot be reproduced or shared without Fair Isaac Corporation’s express consent.

Credit Scoring and Credit Control Conference XVIFrederico Diniz

Analytics Consultant, FICO

Stress Testing: Evaluation of Different Approaches for Modelling Credit Portfolios


Agenda

• Stress Testing Overview

• Introduction

• Regulations and Requirements

• Modelling Framework and Structure

• Modelling Approaches

• Top Down

• Bottom-Up

• Modelling Methodologies

• Scenario Test

• Simulation Methods

• Outcomes and Benefits


Stress Testing Introduction

Stress Testing - Simulations of unexpected events and losses aiming toassess the robustness of a financial institution under rare, but not unrealisticcircumstances.

The Basel Committee on Banking Supervision (BCBS) consider StressTesting to be a critical element of risk management for banks and a core toolfor regulatorsand macroprudential authorities.

• Banks (Micro) : identify vulnerabilities, corroborate that the bank maintainsa robust capital position, provide insights to the bank’s decision makersabout their resilience to economic recessions.

• Regulators (Macro): identify vulnerabilities of the economic system as awhole, support macro-economic scenarios and give indications of potentialfragilities to correct them ahead of an economic recession.


Stress Testing Regulations and Requirements


BCBS is constantly updating andenhancing regulatory policies that willsupport and improve performance ofthe global economy and prepare foreventual economic crises

Each national authority has the responsibilityto interpret the guidelines provided by BCBSand apply them with the necessaryadjustments to protect the regional economy

Stress Testing Regulations and Requirements


Stress Testing Overview

Regulations and Requirements - BCBS

Pillar I – Minimum Capital Requirements;

“An IRB bank must have in place sound stress testing processes for use in the assessment of capital adequacy. Stress

testing must involve identifying possible events or future changes in economic conditions that could have unfavourable

effects on a bank’s credit exposures and assessment of the bank’s ability to withstand such changes. Examples of

scenarios that could be used are:

1. economic or industry downturns;

2. market-risk events; and

3. liquidity conditions.”

Pillar II – Supervisory Review Process;

“National supervisors may wish to issue guidance to their banks on how the tests to be used for this purpose should be

designed, bearing in mind conditions in their jurisdiction. The results of the stress test may indicate no difference in the

capital calculated under the IRB rules described in this section of this Framework if the bank already uses such an

approach for its internal rating purposes.(…)”


Stress Testing Modelling Structure

Early Recovery

Full Recession

Late Recovery

Early Recession

Macro Models

Default Rate

Recovery Rate

Exposure at Default

Bank’ Provision

(ECL)

Regulatory Capital

Management / Risk Appetite

Economic CycleRecession

Impact Affected Parameters Outcome

DataSimulation

and Optimization

Scenario Reporting and Selection

Strategy and Refinement


Stress Testing - Principles

Clear Objectives - Relevant high-level objectives should be aligned with the

bank’s risk appetite and risk management framework.

Governance - specify the roles and responsibilities of senior management,

oversight bodies and those responsible for the ongoing operation of the stress

testing framework.

Risk Management Tool - Results should be reported to the senior management on

a regular basis, at relevant levels of aggregation. The reports should include the

main modelling and scenario assumptions as well as any significant limitations.

Capture Material and Relevant Risks - Comprehensive assessment of risks,

including both on- and off-balance sheet exposures, earnings vulnerabilities,

operational risks, and factors that affect the solvency or liquidity position.


Stress Testing - Principles

Resources and Organisational Structures - Governance processes should ensure the

adequacy of resourcing for stress testing, including ensuring that the resources have

the appropriate skill sets to execute the framework.

Infrastructure - In order for risks to be identified and the results of stress tests to be

reliable, the data used should be accurate and complete, and available at a

sufficiently granular level and in a timely manner.

Models and methodologies should fit for purpose - The models and

methodologies used to derive stress estimates and impacts should fit the purpose and

intended use of the stress tests.

Model Supervision and Review; improving the reliability – Assess the limitations,

identify areas where the stress testing approach should be improved and ensure that

the stress test results are being used in a consistent way.

Communication - This will help to reduce the risk that market participants draw ill-

informed conclusions about the resilience of banks with differing or negative results.


Stress Testing Modelling – Model Design

Stress Testing

ApproachBottom-Up

Top-Down

MethodologyScenario Test

Historical Data

Hypothetical Data

Simulation Methods

Selecting the best Approach

• Who is conducting the stress test exercise?

• Who is responsible for developing the assumptions/scenarios?

• Data Quality/Granularity (Link to the Principles).

• Will the outcome be used for Benchmarking?

• Are my homogeneous portfolios easily identified?

• Size of the Institution.



Top-down Modelling

• Comparability of results;

• Less accurate – especially when carried out onaggregated systemwide data.

• Depends critically on data availability by national authorities;

• Relatively simpler but their accuracy tends to be lower;

Regulators (central banks or supervisory authorities) apply the economic shock – either to individual institutions data or to an aggregated banking system level – and analyse its results on the banking system as a whole.

Key Notes: Example of Usage :

• Central Banks and Regulator aiming riskidentification and communicationpurposes



Bottom-Up ModellingRegulators define the macroeconomic impact (or a set of shocks), let the institutionsevaluate its own impact on their balance sheets and then consolidate the bank-levelresults aiming to report the overall outcome.

• Banks creates shock on its own portfolio(accuracy);

• Better data simulations, possibility of loan-level models and sub-portfolios;

• Hard to compare across industry differentmethodologies and modelling assumptions;

• Interdependencies across institutions.

• May significantly underestimate the true totalrisk.

• EBA Guidelines on Institutions’ Stress

Testing - EBA/GL/2018/04

• PRA Stress testing the UK banking system: 2019 guidance for participating banks and building societies

Example of Usage:Key Notes:



Scenario Test

Historical Data

Hypothetical Data

• The use of historical data to assess the common variables to understand its relationship

with historical recessions.

• Its based on past experience as embodied in historical data, which may not be relevant

to the next stress event. Moreover

• Covers relatively small or incomplete time periods, missing information on important

drivers of loss.

• Data inputs depend on hypothetical stress scenario.

• A risk driver which was not incorporated in the model can be included in a hypothetical

stress scenario.

• Hypothetical scenarios are more flexible to include information which were not tested

before.

• A large number of scenarios is generated through data simulation methods (I.e. Monte

Carlo Simulations)

• Risk factors are set to their stressed values and the remaining risk factors are drawn

from the joint risk factor distribution.

Simulation Methods


© 2019 Fair Isaac Corporation. Confidential.

This presentation is provided for the recipient only and cannot be reproduced or shared without Fair Isaac Corporation’s express consent.

Thank You!

Frederico Diniz

Phone: +44 (0) 786 040 5138Email: [email protected]

stress testing: evaluation of different approaches for ... · stress testing introduction stress...

Documents